JPH0137062B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to a method for driving a matrix circuit using a thin film semiconductor, and particularly to a method for driving a matrix circuit in a reading or display device.

[Prior art] In recent years, hydrogenated amorphous silicon (a-Si:H) and sintered cadmium sulfide-cadmium selenide (CdS-CdSe) have been used for long image sensors for facsimiles, two-dimensional liquid crystal displays, etc. Thin film semiconductors are being used in bodies, etc., and are attracting attention.

Thin film semiconductors can be easily deposited on transparent substrates by methods such as glow discharge, reactive sputtering, and evaporation, if desired, and can also be fabricated by conventional photolithography processes into photodiodes, photodiodes, and photodiodes. It can be processed into arrays such as conductive photo sensors and field effect transistors. Therefore, it has the feature that it is possible to manufacture long and large-area reading and display devices at low cost, which was not possible with conventional crystalline semiconductors.

For reading and display devices, circuit simplification, 2
For the purpose of dimensionalization, matrix circuits are usually used. Therefore, hereinafter, a long image sensor will be taken as an example, and its matrix circuit will be explained.

1 and 2 are matrix circuit diagrams of a long image sensor. In both figures, there are n unit elements e of thin film semiconductors (however, in the drawings, n=
4) They are connected to form one block, and m blocks are connected to form an array. For convenience of explanation, the unit element will be referred to as eij, the suffix i will be the block number, and the suffix j will be the order within the block to which the unit element belongs. Therefore, 1≦i≦m and 1≦j≦n.

In FIG. 1, one terminal of the unit element is commonly connected to each block, and the other terminal is connected to each unit element having the same suffix j in each block (ei ₁ to ei ₄ ; 1≦i≦4). Connected to line 1 ₁ ~ L ₄ . Lines l ₁ -l ₄ are each connected to a switch array of switches 1 - 4 and are connected to ground or to the input terminal of amplifier 5 depending on the operation of switches 1 - 4 .

A voltage Vi (i is the block number; 1≦i≦m) is applied to each unit element eij for each block.
When Vi is applied, the unit element ei ₁ of the block
~ei ₄ becomes active. Here, the image
Since we are taking a sensor as an example, a current corresponding to the intensity of the incident light flows through the unit elements ei ₁ to _{ei 4} that have become active, and this current is sequentially sent to the amplifier 5 by the operation of the switch array. input and amplified. Furthermore, since the voltages V ₁ to Vm are sequentially applied as shown in the timing chart of FIG. 3, currents corresponding to the intensity of incident light flowing through all unit elements e are sequentially input to the amplifier 5. It happens.

On the other hand, in _the matrix circuit shown in _FIG . It is connected. The shift register 10 sequentially outputs stored contents as a time-series signal every time a shift pulse (not shown) is input. Since the voltages V ₁ to Vm are applied as shown in FIG. 3, in the matrix circuit in FIG. 2 as well as in the matrix circuit in FIG. 1, the current corresponding to the incident light intensity is This is obtained sequentially over the array of unit elements e.

By the way, the operation of the entire array of unit elements eij is
In order to complete the process in Ta seconds, each unit element eij needs to be in a normal operating state after Ta/m seconds at the latest after the voltage Vi is applied. For example, Ta
= 10 msec, m = 64, Ta/m = 156 μsec
It is. However, although there is a margin of 156 μsec in calculations, in reality, due to various constraints, the margin is only about 50 μsec.

Figure 4 shows a unit element eij with a gap length of 10μ with ohmic contact between the electrodes in the n ⁺ layer.
2 is a graph showing a change in current over time immediately after applying a voltage of 10 V using a coplanar type photoconductive photo sensor of M. The horizontal axis shows time (μsec) and the vertical axis shows current (A).

Figure 4 a shows illuminance of 100 (lx), Figure 4 b shows illuminance of 10
(lx), and Figure 4 c is dark.
Each case of the state is shown.

As is clear from these graphs, the voltage 10
A large current flows immediately after V is applied, but about
After 200μsec has passed, it is in a steady state. but,
In the case of 10 (lx) {Fig. 4b} and in the dark state {Fig. 4c}, the current immediately after voltage application is particularly large compared to the current in the steady state. for that,
In steady state, the current at 100 (lx) is 10
(lx), whereas it is about 5 times the current in the case of (lx).
Immediately after voltage application, it is only 2.3 times as large, indicating that it is becoming difficult to distinguish between light intensities.
In other words, conventional image sensors tend to malfunction when reading documents.

To solve this problem, either increase the operation time Ta of the entire array of unit elements eij, or
Alternatively, a method can be considered in which the number n of unit elements in a block is increased to increase the number of switches 1 to 4 in FIG. 1 and the number of amplifiers 6 to 9 in FIG. 2.

However, these methods degrade device performance and increase cost, so they are not ideal solutions.

[Object of the Invention] The present invention has been made in view of the above-mentioned conventional problems, and its purpose is to provide a matrix circuit driving method that is less likely to cause malfunctions and can realize a high-speed, low-cost matrix circuit. There is a particular thing.

[Summary of the Invention] In order to achieve the above object, the present invention provides a matrix circuit driving method in which a plurality of thin film semiconductor unit elements that become active by applying a voltage are connected, and a voltage is simultaneously applied to each unit element. The present invention has a plurality of blocks configured as shown in FIG. In a matrix circuit driving method in which a period in which no block is in an active state is provided between each period in which a voltage is applied to make any one of the blocks active, The present invention is characterized in that a voltage is applied to the one block during a period in which a voltage for activating blocks other than the one block is not applied.

[Embodiments of the Invention] Before describing embodiments of the present invention, the theoretical basis of how the matrix circuit driving method according to the present invention is possible will be explained first.

Figure 5 is a graph showing the pulse frequency dependence of current Ip at 50 μsec after the pulse rises when a repeated pulse voltage (in this case, voltage 10 V, pulse duty 50%) is applied to a unit element of a thin film semiconductor. It is. When curve 11 has an illuminance of 100 (lx),
Each curve 12 represents the case of 10 (lx).

As can be seen from this graph, in both cases of 100 (lx) and 10 (lx), as the frequency of the repetitive pulse increases, the current Ip tends to decrease, especially in the case of 10 (lx). The decline has become significant. Therefore, in the high frequency region, 100
The current Ip in the case of (lx) is approximately 4.3 times the current Ip in the case of 10 (lx), which is close to the steady state case (approximately 5 times). This phenomenon will be discussed theoretically below.

When an electrode is attached to a relatively high-resistance semiconductor and a strong electric field is applied, carriers (eg, electrons) are generally injected from the electrode, forming a space charge inside the semiconductor. The current flowing through a semiconductor is determined by this space charge, and such current is called a space charge limited current (Space Charge Limited Current).
Current (hereinafter referred to as SCLC). In steady state, the magnitude I of SCL is given by the following equation.

I=KV ² θμ/4πL ³ ×10 [A/cm ² ] ………(1) However, K is the dielectric constant, V is the applied voltage, μ is the mobility of the semiconductor, L is the distance between the electrodes, and θ is It is the ratio (Nc/Nt) between the carrier concentration Nc in the conduction band of a semiconductor and the carrier concentration t in the shallow trap level where the carriers do not serve as recombination centers.

However, the state immediately after an electric field is applied to the semiconductor is not a steady state, and the carriers injected from the electrodes hardly fall to the trap level. This initial state is schematically shown in FIG. 6a.

In the figure, a voltage has just been applied to both ends of the semiconductor 13, that is, the positive electrode side 14 and the negative electrode side 15, and a large number of holes 18 are generated on the brass electrode side 14. Although a shallow trap level 16 exists in the semiconductor 13, the electrons 17 injected from the negative electrode side 15 have not yet fallen into the shallow trap level 16 in this initial state.

In such an initial state, Nc is sufficiently larger than Nt, so the value of θ becomes large, and therefore SCLC also becomes large.

However, as time passes, the annihilation due to the coupling between the electrons 17 and the holes 18 and the supply from the electrodes approach an equilibrium state, and the electrons 17 fall into the shallow trap level 16 (state 19), or the electrons 17 fall into the shallow trap level 16 (state 19), or The electrons are excited from the level 16 to the conduction band again, and the electron concentration becomes balanced between the conduction band and the trap level 16. for that,
The value of θ approaches a constant value that is smaller than that in the initial state, and accordingly, SCLC is also smaller than the initial value and approaches a constant value. With this way of thinking,
The phenomenon in which each of the graphs in FIG. 4 shows a large current value initially and then gradually settles down to a steady value can be explained.

In the case of FIGS. 4a and 4b, that is, when the semiconductor is irradiated with light, the situation becomes complicated, but it can be approximately expressed by the following equation.

I=qμNc(F)V/L+KV ² θμ/4πL ³ ………(2) Here, q is the electric charge, and Nc(F) is the conduction band value corresponding to the incident light intensity F in the state where no electric field is applied. It is the electron concentration.

The first term in equation (2) represents a current that changes depending on the incident light intensity F, and the second term represents SCLC. That is, the difference in current value in the steady state in each graph of FIG. 4a and b reflects the difference in current due to the first term.

However, as already mentioned, in the initial state immediately after voltage application, the second term becomes sufficiently large, so
The difference in current value due to the difference in the incident light intensity F in the first term is less likely to be reflected. In other words, it can be considered that the dependence of the current on light intensity becomes smaller immediately after voltage application. This resulted in malfunctions of conventional image sensors and the like.

By the way, as shown in Fig. 5, when a pulse voltage is applied repeatedly, the
The electric charge Ip at 50 μsec decreases depending on the pulse frequency, and a phenomenon is observed in which the dependence of the current Ip on light intensity increases, especially in a certain high frequency range. This phenomenon can be understood from the above idea as follows.

That is, when a repetitive pulse voltage of a certain high frequency is applied, there is no room for electrons to escape from the shallow trap level, so that the electrons always remain in the shallow trap level. for that,
θ in the second term of equation (2) is not sufficiently large in the initial state, and the current Ip decreases accordingly, and the current in the first term is largely reflected in the current Ip. That is, the difference in light intensity F is sufficiently reflected in the current Ip.

Note that the characteristics of the current transient response after voltage application as described above are considered to be remarkable in thin film semiconductors, which are known to have many shallow trap levels.

Now, based on the experimental results and theoretical considerations explained above, we will return to the matrix circuit shown in FIGS. 1 and 2.

In the matrix circuit shown in FIGS. 1 and 2, a voltage Vi is applied to each block at the timing shown in FIG. In the timing chart shown in Figure 3, each block has a voltage
During each period activated by the application of Vi,
It is possible to provide periods during which no block is active. Then, if voltage is applied to all blocks during this provided period, any one
For the block, a pulse voltage of a specific period is applied before it becomes active. A timing chart of this voltage Vi is shown in FIG.

FIG. 7 is a timing chart of the applied voltage Vi (1≦i≦5) when the number of blocks in the matrix circuit shown in FIGS. 1 and 2 is m=5, and is a timing chart for driving the matrix circuit according to the present invention. 1 is a first embodiment of the method;

Taking the voltage _V4 applied to the fourth block as an example, a repeated pulse voltage is applied before the period _T4 when the fourth block becomes active. These repetitive pulse voltages can be obtained by applying voltages V 1 to _{V 5 during} periods P ₁ to P ₄ other than periods T ₁ to T ₃ during which the first to third blocks are active. . This situation is, of course, common to all blocks, not just the fourth block.

By applying such a voltage Vi, the dependence of the photocurrent of the unit element eij on the light intensity increases.
Already explained.

However, in actual operation, it is desirable that the ratio Ti/(Ti+Pi) of the active state period of each block be large. In other words, it is desirable that the duty of the repetitive pulse be small. The maximum value of this duty is Pi/(Ti+Pi), but as shown in FIG. 8, it can be seen that even if the duty is reduced, the light intensity dependence of the current Ip does not decrease much.
Therefore, characteristics can be improved without substantially reducing the active state ratio. However, in the graph shown in Figure 8, the duty of the repetitive voltage pulse is plotted on the horizontal axis, and the current Ip (100) at 100 (lx) is plotted on the vertical axis.
and the current Ip(10) at 10(lx) [Ip(100)/Ip
(10)] is taken.

Next, problems and solutions when applying the timing chart of FIG. 7, which is an embodiment of the present invention, to FIGS. 1 and 2 will be explained.

When all block voltages are applied at the timing shown in FIG. 7, lines li to l ₄ in period Pi
A large current flows through the If this current is input to amplifier 5 or amplifiers 6 to 9, there is a possibility that the dynamic range of the amplifier will be exceeded, and the characteristics of the amplifier will be adversely affected. To solve this problem, in the matrix circuit shown in Figure 1, the line is
Switches 1 to 4 may be operated so that l ₁ to l ₄ are all grounded.

Further, in the matrix circuit shown in FIG. 2, the problem can be solved by providing a bypass circuit before the amplifiers 6 to 9.

FIG. 9 shows an example of a bypass circuit in which a Schottky diode 20 is connected in front of the amplifier 6 in FIG. 2 to release the large current in line _l1 . Of course, the other amplifiers 7, 8, and 9 are also provided with similar bypass circuits. shot key
The diode 20, as shown in FIG.
Even in the forward direction, when the voltage V is small, almost no current I flows, and when the voltage V increases to a certain extent, the resistance rapidly decreases and a large current I flows. By utilizing this characteristic, for example, even if a large current flows through line l ₁ (the same applies to the other lines l ₂ to l ₄ ) and the amplifier 6 becomes saturated, the resistance of the Schottky diode 20 becomes small, so the amplifier 6 can be prevented from increasing the input voltage. Conversely, when a certain block becomes active and the current on line _l1 decreases, the input voltage of the amplifier is low enough that the Schottky diode 20 is in a high resistance state, and the line
The current of _l1 is input directly to amplifier 6.

The above explanation was about the first embodiment of the present invention in which the voltage Vi is applied at the timing shown in FIG. 7, but it is not necessarily necessary to always repeatedly apply voltage pulses to all blocks in this way. do not have. It has been found that the effect is sufficiently increased if a suitable number k of voltage pulses are applied before any given block becomes active.

In Figure 11, the horizontal axis is the number of pulses k, and the vertical axis is the current Ip.
This is a graph showing the relationship between the number of pulses k and the current Ip. When curve 21 is 100 (lx), curve 22
shows the case of 10 (lx). As shown in this graph, whether the number k of voltage pulses applied in advance is 10 or 5,
It can be seen that the ratio [Ip(100)/Ip(10)] between the current Ip(100) in the case of 100(lx) and the current Ip(10) in the case of 10(lx) is a sufficiently large value.

Hereinafter, specific examples of the matrix circuit schematically shown in FIGS. 1 and 2 will be shown, and using these specific examples, a first embodiment of the matrix circuit driving method according to the present invention shown in FIG. 7 will be explained. This will be explained in more detail.

The circuit diagram shown in FIG. 12 is a specific example of the matrix circuit shown in FIG. However, the first
In Figure 2, the unit element eij of the thin film semiconductor is 32
Each block forms one block, and 64 blocks are connected to form the matrix section 23. Therefore, in this case m=64 and n=32.

The matrix portion 23 is manufactured in the next step. First, a cleaned glass substrate (manufactured by Corning, 7059 glass) was placed on the anode in a glow discharge device, and then the vacuum inside the device was set to 10 ⁻⁶ Torr. Next, add 10SCCM of high purity monosilane gas (SiH ₄ )
(Standard cc/min) and phosphine gas (PH ₃ ) diluted to 10 ppm with high purity hydrogen gas (H ₂ ) was flowed into the apparatus at a flow rate of 5 SCCM. At this time, the pressure inside the device was maintained at 0.1 Torr. Next, a high frequency of 1356 MHz was applied between the parallel plate electrodes to generate a glow discharge. and,
An a-Si layer of about 7000 Å was deposited on a glass substrate.
At this time, the glass substrate was kept at 200°C. Subsequently, SiH ₄ was flowed into the device at a flow rate of 2 SCCM, and PH ₃ gas diluted to 1000 ppm with high-purity hydrogen gas (H ₂ ) was flowed into the device at a flow rate of 10 SCCM, and glow discharge was performed in the same manner. A low-resistance n ⁺ layer of about 1000 Å was deposited on the -Si layer.

Further, Al was deposited to a thickness of about 2000 Å on the n ⁺ layer by vacuum evaporation, and then the Al deposited layer except for the portions that would become the electrodes and underlying wiring was removed using a normal photolithography technique.

Next, the exposed n ⁺ layer is removed by dry etching using the Al pattern as a mask to complete the unit device eij.

Next, unit elements eij are wired in a matrix. First, polyimide resin (trade name: PIQ = polyimide isoindoquinazolinedione) was applied and baked, and then a contact hole was made to establish electrical connection with the upper layer wiring using a photolithography process. Aluminum Al was then deposited to a thickness of about 5000 Å using a vacuum evaporation method, and upper layer wiring was formed again using a photolithography process.

Matrix part 2 formed by the above steps
3 is a common electrode side drive section 24 (hereinafter referred to as common section 24) that supplies an applied voltage Vi, and a unit element.
They are each connected to an individual electrode side drive section 25 (hereinafter referred to as an individual section 25) which inputs the photocurrent of eij and outputs it as a time-series signal.

The common section 24 is configured as follows. The parallel output terminals of the shift register 26 (here, 64 bits) are connected to the input terminal of the inverter INi (1≦i≦64, the same applies hereinafter) and the gate terminal of the transistor _TRi1 , and the output terminal of the inverter INi is connected to the transistor TRi1. Connected to the gate terminal of ₀ . Also, the positive terminal of the DC power supply 27 is the source terminal (or drain terminal) of the transistor TRi ₁ .
The negative terminal is grounded and connected to the drain terminal (or source terminal) of the transistor TRi ₀ . The drain terminal (or source terminal) of transistor TRi ₁ and the transistor
Both source terminals (or drain terminals) of TRi ₀ are connected to a common terminal of unit elements eij of the matrix section 23 to supply an applied voltage Vi.

Next, the configuration of the individual section 25 will be shown. transistor
The source terminals (or drain terminals) of TRAj ₀ (1≧j≧32, the same applies hereinafter) are connected to the lines lj of the matrix section 23, and are also connected to the source terminals (or drain terminals) of the transistor TRAji. Shift register 28 (here
The parallel output terminal of the 32-bit) is connected to the input terminal of the inverter INj, and the
Connected to the gate terminal of TRAj ₁ . The drain terminal (or source terminal) of the transistor TRAj ₁ is connected to the input terminal of the amplifier 29, and the drain terminal (or source terminal) of the transistor TRAj ₀ is grounded.

A case in which this matrix circuit is operated by the driving method according to the present invention will be described next.

First, at the timing shown in FIG.
In order to apply the pulse signal S 1 shown in FIG. 13, the pulse signal S ₁ shown in FIG. 13 is input to the shift register 26 of the common section 24 and sequentially shifted using a 50 KHz shift pulse.

In this example, the period Ta of the pulse signal _S1 is 5.12
msec, pulse width △P of repeated pulse is 20μsec,
The interval between repeated pulses or the period during which a voltage is applied to activate an arbitrary block ΔT is 60 μsec.

Taking the first block (suffix i=1) in FIG. 12 as an example, the shift register 26
If the content of _R1 is high level, the transistor
TR ₁₁ is turned on, and the voltage V of the DC power supply 27 is supplied to the first block as the applied voltage _V1 . When the content of R ₁ goes low level, transistor TR ₁₁
is turned off, and conversely, a high level is applied to the gate terminal of the transistor _TR10 by the inverter _IN1 . Therefore, the transistor _TR10 is turned on, and the common terminal of the first block is grounded and the voltage
V ₁ =0. Therefore, by passing the pulse signal S ₁ shown in FIG. 13 through R ₁ of the shift register 26, the applied voltage V ₁ can be changed at the same timing. This situation is the same for the applied voltages V _{2 to V 64} of the other blocks, apart from the time delay.
The applied voltage Vi can be obtained at the timing shown in the figure.

On the other hand, in the individual section 25, the unit element is
An operation is performed in which the photocurrents ei ₁ to _{ei 32} are sequentially sent to the amplifier 29.

For this purpose, a pulse signal _S2 shown in FIG. 14b is input to the shift register 28, and shifted by a 1 MHz shift pulse. However, the 14th
Figure a is the pulse signal _S1 in Figure 13.

In this embodiment, the period △P+△TA from the time when the pulse of pulse signal _S1 rises until the pulse of pulse signal _S2 rises is 20μsec+28μsec=
The pulse width ΔPe of the pulse signal _S2 is 1 μsec. Therefore, the period ΔTB during which the pulse signal S ₂ shifts from SR ₁ to SR ₃₂ of the shift register 28
is 32μsec.

As an example, the pulse signal shown in FIG.
Voltage Vi is applied during period △T of S ₁ and unit element ei ₁ ~
Consider the case where ei ₃₂ becomes active. When ΔTA=28 μsec has elapsed from the start of period ΔT, pulse signal S ₂ is input to SR ₁ of shift register 28 and becomes high level. Transistor for that
TRA ₁₁ is turned on, and the photocurrent flowing through the unit element ei ₁ is input to the amplifier 29. Thereafter, the high level is shifted from SR ₂ to _{SR 32} by a 1 MHz shift pulse, and accordingly, the photocurrents of the unit elements ei ₂ to _{ei 32} are sequentially input to the amplifier 29, and a time series signal S ₀ is obtained. . At this time, during the period ΔP during which the voltage Vi is a repetitive pulse, the contents of the shift register 28 are all at a low level, so a high level is applied to the gate terminal of the transistor TRAj ₀ via the inverter INVj, and the transistor TRAj ₀ turns ON and line 1j is grounded.

FIG. 15a is an output waveform diagram of the amplifier 29 when the conventional voltage pulse shown in FIG. 3 is applied to the matrix circuit in FIG.
Figure b shows amplifier 2 using an embodiment of the present invention.
9 is an output waveform diagram of No. 9. Curve 30 shows the case of 100 (lx), and curve 31 shows the case of 10 (lx).

In Fig. 15a, even though the illuminance is uniform, the magnitude of the output signal is different at the beginning and end of the unit element within one block, and furthermore, the magnitude of the output signal is different at the beginning and end of the unit element within one block.
The ratio of the output signal in the case of (lx) to that in the case of 10(lx) is becoming smaller. In contrast, a significant improvement is observed in FIG. 15b.

FIG. 16 is a specific circuit diagram of the matrix circuit in FIG. 2. In the figure, matrix section 23
Since the common section 24 and the common section 24 are the same circuits as in FIG. 12, and the pulse signal _S1 input to the shift register 26 is also the same, the explanation will be omitted and only the individual section 25 will be explained.

One terminal of the Schottky diode Dj (1≦j≦32 or less) is connected to the line lj so that the forward direction occurs when the line lj becomes a high potential, and the other terminal is grounded. line lj
Furthermore, they are each connected to the input terminals of an amplifier AMPj, and the output terminals of the amplifier AMPj are respectively connected to parallel input terminals of a shift register 33 via a sample and hold circuit 32.

FIG. 17a shows the pulse signal S ₁ input to the shift register 26. Therefore, it is assumed that a voltage Vi for activating the unit elements ei ₁ to ei ₃₂ is applied to the pulse interval ΔT of the pulse signal S ₁ in FIG. 17a. At this time, the photocurrents flowing through the unit elements ei ₁ to _{ei 32} are amplified by the amplifiers AMP ₁ to _{AMP 32} and input to the sample and hold circuit 32 .
However, the sample and hold circuit 32 does not hold the signal from the amplifier AMPj unless the hold signal _S3 in FIG. 17b is input.

As shown in FIG. 17b, the hold signal S ₃
is input to the sample hold circuit 32 at the end of the period ΔT, and the amplifier AMP ₁ ~
The output of the AMP ₃₂ is held in the sample hold circuit 32 and stored in the shift register 33. Subsequently, a 1MHz shift pulse as shown in FIG. 17c is applied to the shift register 33 for a period ΔTC=
The signal is applied for 32μsec, and the stored contents are output as a time series signal _S0 from the serial output terminal.

FIG. 18a is a waveform diagram of the time-series signal S ₀ when voltage Vi with the timing shown in FIG. 3 is applied, and FIG. 18b is a waveform diagram of the driving method according to the present invention shown in FIG. FIG. 3 is a waveform diagram of a time-series signal S ₀ obtained when using the method. Curve 34 shows the case of 100 (lx), and curve 35 shows the case of 10 (lx).

In the case of the matrix circuit shown in FIG.
Since the signal is extracted by the sample-and-hold circuit at the end of the period ΔT in which the unit element eij is in the active state, each unit element is in a stable state and there is no difference in output value before and after. However, as shown in FIG. 18a, the output signal ratio in the case of 100 (lx) to that in the case of 10 (lx) remains small. On the other hand, when the driving method according to the present invention is used, there is a significant improvement as shown in FIG. 18b.

It should be noted that although the explanation has been given using a long image sensor as an example, the driving method according to the present invention is not limited to this and can be applied to driving other thin film semiconductor devices.

For example, LCD (liquid crystal display), ECD
TFT used for (electrochromic display) etc.
(Thin film transistor) It can also be applied to two-dimensional devices.

[Effects of the Invention] As explained in detail above, the matrix circuit driving method according to the present invention has a simple configuration and is capable of improving the activity efficiency of unit elements of thin film semiconductors. This has the great effect of realizing a circuit.

[Brief explanation of drawings]

Fig. 1 is a schematic circuit diagram showing a first example of a matrix circuit, Fig. 2 is a schematic circuit diagram showing a second example of a matrix circuit, and Fig. 3 shows the application timing of applied voltages V ₁ to Vm. The timing chart shown in Figure 4 shows the time characteristics of the photocurrent of a thin film semiconductor. Figure 4a is a characteristic curve diagram for the case of 100 (lx), and Figure 4b is the characteristic curve diagram for the case of 10 (lx). Characteristic curve diagram, Fig. 4c
is a characteristic curve diagram in the dark state, Figure 5 is a frequency characteristic curve diagram of photocurrent of a thin film semiconductor, Figure 6 a is a schematic diagram showing the initial state when voltage is applied to the semiconductor, and Figure 6 b is a steady state diagram. A schematic diagram showing the state, FIG. 7 is a timing chart of the applied voltage Vi showing an example of the present invention, and FIG. 8 is a characteristic curve showing the change in light intensity dependence when the duty of the repetitive pulse is changed. Figure 9 is a partial wiring diagram when a diode is added to the matrix circuit in Figure 2,
FIG. 10 is a voltage-current characteristic curve diagram of the diode in FIG. 9, FIG. 11 is a characteristic curve diagram showing the relationship between the number of pulses and photocurrent in a thin film semiconductor, and FIG.
The figure is a circuit diagram showing the specific configuration of the matrix circuit in Figure 1, Figure 13 is a waveform diagram of pulse signal S ₁ in Figure 12, Figure 14a is a waveform diagram of pulse signal S ₁ , and Figure 14b is 15 is a waveform diagram of the pulse signal S ₂ at the timing of the pulse signal S ₁ , FIG. 15a is an output signal waveform diagram obtained by the conventional driving method in the matrix circuit in FIG. A diagram of the output signal waveform obtained by the driving method, FIG. 16 is a concrete circuit diagram in which a diode is added to the matrix circuit in FIG. 2, and FIG. 17 a, b, and c are pulse signals S ₁ , S ₃ , _S4 waveform diagram, FIG. 18a is an output signal waveform diagram obtained by the conventional driving method in the matrix circuit in FIG. 16, and FIG. 18b is an output signal waveform diagram obtained by the driving method of the present invention. FIG. e (1≦i≦m, 1≦j≦n)... Thin film semiconductor unit element, 23... Matrix section, 24... Common electrode side drive section, 25... Individual electrode side drive section.

Claims (1)

[Scope of Claims] 1. A device comprising a plurality of blocks configured such that a plurality of thin film semiconductor unit elements that become active by applying a voltage are connected and a voltage is simultaneously applied to each unit element, By sequentially applying a voltage to the blocks, the plurality of unit elements are sequentially activated block by block, and during each period in which each block is activated by voltage application, no block is activated. In the matrix circuit driving method, the matrix circuit driving method includes a period during which a block other than the one block is activated during a period before a voltage is applied to make any one block among the plurality of blocks active. A method for driving a matrix circuit, characterized in that a voltage is applied to the one block during a period in which a voltage for setting the block is not applied.